Paper Title
A Comparative Analysis of Machine Learning Models for Pressure Reconstruction in Particle Image Velocimetry (PIV)

Abstract
The experimental method known as Particle Image Velocimetry (PIV) is frequently used to assess fluid flow. A significant problem in PIV is pressure reconstruction from velocity field data. For pressure reconstruction in PIV, machine learning (ML) based algorithms have recently been put out as an alternative to conventional numerical methods. In this paper, multiple ML-based regressors for pressure reconstruction in PIV applications are quantitatively analyzed. The performance of several ML-based regressors, including Support Vector Regression (SVR) and Random Forest (RF), was evaluated using a dataset of artificially generated velocity fields. The performance indicators utilized for comparison are mean absolute error (MAE), root mean square error (RMSE), and coefficient of determination (R2). The findings show that, in comparison to conventional numerical techniques, ML-based regressors can produce accurate pressure reconstruction at a much lower computing cost. Gradient Boosting was discovered to perform the best in terms of accuracy and computational efficiency among the numerous ML-based regressors studied.